COVLIAS 3.0: Cloud-based Quantized Hybrid UNet3+ Deep Learning for COVID-19 Lesion Detection in Lung Computed Tomography

Overview

Journal Front Artif Intell

Specialty Biomedical Engineering

Date 2024 Jul 15

PMID 39006802

Authors

Sushant Agarwal

Sanjay Saxena

Alessandro Carriero

Gian Luca Chabert

Gobinath Ravindran

Sudip Paul

John R Laird

Deepak Garg

Mostafa Fatemi

Lopamudra Mohanty

Arun K Dubey

Rajesh Singh

Mostafa M Fouda

Narpinder Singh

Subbaram Naidu

Klaudija Viskovic

Melita Kukuljan

Manudeep K Kalra

Luca Saba

Jasjit S Suri

Affiliations

Soon will be listed here.

Abstract

Background And Novelty: When RT-PCR is ineffective in early diagnosis and understanding of COVID-19 severity, Computed Tomography (CT) scans are needed for COVID diagnosis, especially in patients having high ground-glass opacities, consolidations, and crazy paving. Radiologists find the manual method for lesion detection in CT very challenging and tedious. Previously solo deep learning (SDL) was tried but they had low to moderate-level performance. This study presents two new cloud-based quantized deep learning UNet3+ hybrid (HDL) models, which incorporated full-scale skip connections to enhance and improve the detections.

Methodology: Annotations from expert radiologists were used to train one SDL (UNet3+), and two HDL models, namely, VGG-UNet3+ and ResNet-UNet3+. For accuracy, 5-fold cross-validation protocols, training on 3,500 CT scans, and testing on unseen 500 CT scans were adopted in the cloud framework. Two kinds of loss functions were used: Dice Similarity (DS) and binary cross-entropy (BCE). Performance was evaluated using (i) Area error, (ii) DS, (iii) Jaccard Index, (iii) Bland-Altman, and (iv) Correlation plots.

Results: Among the two HDL models, ResNet-UNet3+ was superior to UNet3+ by 17 and 10% for Dice and BCE loss. The models were further compressed using quantization showing a percentage size reduction of 66.76, 36.64, and 46.23%, respectively, for UNet3+, VGG-UNet3+, and ResNet-UNet3+. Its stability and reliability were proved by statistical tests such as the Mann-Whitney, Paired -Test, Wilcoxon test, and Friedman test all of which had a < 0.001.

Conclusion: Full-scale skip connections of UNet3+ with VGG and ResNet in HDL framework proved the hypothesis showing powerful results improving the detection accuracy of COVID-19.

References

Ma J, Wang Y, An X, Ge C, Yu Z, Chen J . Toward data-efficient learning: A benchmark for COVID-19 CT lung and infection segmentation. Med Phys. 2020; 48(3):1197-1210. DOI: 10.1002/mp.14676. View

Saba L, Banchhor S, Suri H, Londhe N, Araki T, Ikeda N . Accurate cloud-based smart IMT measurement, its validation and stroke risk stratification in carotid ultrasound: A web-based point-of-care tool for multicenter clinical trial. Comput Biol Med. 2016; 75:217-34. DOI: 10.1016/j.compbiomed.2016.06.010. View

Jain P, Sharma N, Saba L, Paraskevas K, Kalra M, Johri A . Unseen Artificial Intelligence-Deep Learning Paradigm for Segmentation of Low Atherosclerotic Plaque in Carotid Ultrasound: A Multicenter Cardiovascular Study. Diagnostics (Basel). 2021; 11(12). PMC: 8699942. DOI: 10.3390/diagnostics11122257. View

Dewitte K, Fierens C, Stockl D, Thienpont L . Application of the Bland-Altman plot for interpretation of method-comparison studies: a critical investigation of its practice. Clin Chem. 2002; 48(5):799-801; author reply 801-2. View

Wu X, Hui H, Niu M, Li L, Wang L, He B . Deep learning-based multi-view fusion model for screening 2019 novel coronavirus pneumonia: A multicentre study. Eur J Radiol. 2020; 128:109041. PMC: 7198437. DOI: 10.1016/j.ejrad.2020.109041. View

Saba L, Gerosa C, Fanni D, Marongiu F, La Nasa G, Caocci G . Molecular pathways triggered by COVID-19 in different organs: ACE2 receptor-expressing cells under attack? A review. Eur Rev Med Pharmacol Sci. 2020; 24(23):12609-12622. DOI: 10.26355/eurrev_202012_24058. View

Pathak Y, Tiwari A, Stalin S, Singh S, Shukla P . Deep Transfer Learning Based Classification Model for COVID-19 Disease. Ing Rech Biomed. 2020; 43(2):87-92. PMC: 7238986. DOI: 10.1016/j.irbm.2020.05.003. View

Fourcade A, Khonsari R . Deep learning in medical image analysis: A third eye for doctors. J Stomatol Oral Maxillofac Surg. 2019; 120(4):279-288. DOI: 10.1016/j.jormas.2019.06.002. View

Aslan M, Unlersen M, Sabanci K, Durdu A . CNN-based transfer learning-BiLSTM network: A novel approach for COVID-19 infection detection. Appl Soft Comput. 2020; 98:106912. PMC: 7673219. DOI: 10.1016/j.asoc.2020.106912. View

10.

Acharya R, Faust O, Alvin A, Sree S, Molinari F, Saba L . Symptomatic vs. asymptomatic plaque classification in carotid ultrasound. J Med Syst. 2011; 36(3):1861-71. DOI: 10.1007/s10916-010-9645-2. View

11.

Gao S, Cheng M, Zhao K, Zhang X, Yang M, Torr P . Res2Net: A New Multi-Scale Backbone Architecture. IEEE Trans Pattern Anal Mach Intell. 2019; 43(2):652-662. DOI: 10.1109/TPAMI.2019.2938758. View

12.

Ding W, Abdel-Basset M, Hawash H, ELkomy O . MT-nCov-Net: A Multitask Deep-Learning Framework for Efficient Diagnosis of COVID-19 Using Tomography Scans. IEEE Trans Cybern. 2021; 53(2):1285-1298. DOI: 10.1109/TCYB.2021.3123173. View

13.

Agarwal M, Saba L, Gupta S, Johri A, Khanna N, Mavrogeni S . Wilson disease tissue classification and characterization using seven artificial intelligence models embedded with 3D optimization paradigm on a weak training brain magnetic resonance imaging datasets: a supercomputer application. Med Biol Eng Comput. 2021; 59(3):511-533. DOI: 10.1007/s11517-021-02322-0. View

14.

Maniruzzaman M, Rahman M, Al-MehediHasan M, Suri H, Abedin M, El-Baz A . Accurate Diabetes Risk Stratification Using Machine Learning: Role of Missing Value and Outliers. J Med Syst. 2018; 42(5):92. PMC: 5893681. DOI: 10.1007/s10916-018-0940-7. View

15.

Cau R, Bassareo P, Mannelli L, Suri J, Saba L . Imaging in COVID-19-related myocardial injury. Int J Cardiovasc Imaging. 2020; 37(4):1349-1360. PMC: 7676417. DOI: 10.1007/s10554-020-02089-9. View

16.

Gifani P, Shalbaf A, Vafaeezadeh M . Automated detection of COVID-19 using ensemble of transfer learning with deep convolutional neural network based on CT scans. Int J Comput Assist Radiol Surg. 2020; 16(1):115-123. PMC: 7667011. DOI: 10.1007/s11548-020-02286-w. View

17.

Fanni D, Saba L, Demontis R, Gerosa C, Chighine A, Nioi M . Vaccine-induced severe thrombotic thrombocytopenia following COVID-19 vaccination: a report of an autoptic case and review of the literature. Eur Rev Med Pharmacol Sci. 2021; 25(15):5063-5069. DOI: 10.26355/eurrev_202108_26464. View

18.

Basar S, Waheed A, Ali M, Zahid S, Zareei M, Biswal R . An Efficient Defocus Blur Segmentation Scheme Based on Hybrid LTP and PCNN. Sensors (Basel). 2022; 22(7). PMC: 9003284. DOI: 10.3390/s22072724. View

19.

Biswas M, Kuppili V, Saba L, Edla D, Suri H, Sharma A . Deep learning fully convolution network for lumen characterization in diabetic patients using carotid ultrasound: a tool for stroke risk. Med Biol Eng Comput. 2018; 57(2):543-564. DOI: 10.1007/s11517-018-1897-x. View

20.

Hou S, Jia C, Hou M, Fernandes S, Guo J . A Study on Weak Edge Detection of COVID-19's CT Images Based on Histogram Equalization and Improved Canny Algorithm. Comput Math Methods Med. 2021; 2021:5208940. PMC: 8568529. DOI: 10.1155/2021/5208940. View